[
https://issues.apache.org/jira/browse/CASSANDRA-10134?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15211797#comment-15211797
]
Sam Tunnicliffe commented on CASSANDRA-10134:
---------------------------------------------
bq. If we expose "in a shadow round" in some form
I think this is pretty simple to do. Instead of always dropping syns
immediately whenever gossip is disabled, a node currently in a shadow round
could respond with a minimal ack. The node receiving the syn can infer that the
sender is in a shadow round itself by inspecting the syn, as a "real" syn will
never have an empty digest list. So, the syn-receiving node can preserve
current behaviour when the sender is not in a shadow round, but respond with
the minimal ack when it is. When in a shadow round, a node can keep track of
which seeds have replied to its syns with such a minimal ack, then the decision
about exiting the round becomes whether any "genuine" ack was received (only
one is required, as current behaviour) or whether a "shadow" ack was received
from every seed.
A brief experiment with this approach seems to suggest it's viable, Tyler's
dtest passes and startup time for fresh clusters is minimally impacted. This
doesn't add a great deal of complexity, so unless I overlooked something it
seems like a reasonable idea.
> Always require replace_address to replace existing address
> ----------------------------------------------------------
>
> Key: CASSANDRA-10134
> URL: https://issues.apache.org/jira/browse/CASSANDRA-10134
> Project: Cassandra
> Issue Type: Improvement
> Components: Distributed Metadata
> Reporter: Tyler Hobbs
> Assignee: Sam Tunnicliffe
> Fix For: 3.x
>
>
> Normally, when a node is started from a clean state with the same address as
> an existing down node, it will fail to start with an error like this:
> {noformat}
> ERROR [main] 2015-08-19 15:07:51,577 CassandraDaemon.java:554 - Exception
> encountered during startup
> java.lang.RuntimeException: A node with address /127.0.0.3 already exists,
> cancelling join. Use cassandra.replace_address if you want to replace this
> node.
> at
> org.apache.cassandra.service.StorageService.checkForEndpointCollision(StorageService.java:543)
> ~[main/:na]
> at
> org.apache.cassandra.service.StorageService.prepareToJoin(StorageService.java:783)
> ~[main/:na]
> at
> org.apache.cassandra.service.StorageService.initServer(StorageService.java:720)
> ~[main/:na]
> at
> org.apache.cassandra.service.StorageService.initServer(StorageService.java:611)
> ~[main/:na]
> at
> org.apache.cassandra.service.CassandraDaemon.setup(CassandraDaemon.java:378)
> [main/:na]
> at
> org.apache.cassandra.service.CassandraDaemon.activate(CassandraDaemon.java:537)
> [main/:na]
> at
> org.apache.cassandra.service.CassandraDaemon.main(CassandraDaemon.java:626)
> [main/:na]
> {noformat}
> However, if {{auto_bootstrap}} is set to false or the node is in its own seed
> list, it will not throw this error and will start normally. The new node
> then takes over the host ID of the old node (even if the tokens are
> different), and the only message you will see is a warning in the other
> nodes' logs:
> {noformat}
> logger.warn("Changing {}'s host ID from {} to {}", endpoint, storedId,
> hostId);
> {noformat}
> This could cause an operator to accidentally wipe out the token information
> for a down node without replacing it. To fix this, we should check for an
> endpoint collision even if {{auto_bootstrap}} is false or the node is a seed.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
